Dennis R. Helsel

Statistics for Censored Environmental Data Using Minitab and R

• Gebundenes Buch

Produktbeschreibung

This Second Edition solves a current dilemma that occurs across a wide spectrum of environmental science: how to correctly analyze and interpret censored data (data below detection limits). It adapts survival analysis methods and demonstrates their practical applications when studying trace chemicals in air, water, soils, and biota. This edition features new chapters on plotting data with nondetects, multivariate procedures, and software for data with nondetects, as well as expanded and new sections, including a new section on testing censored data for normal distributions. Environmental professionals and upper-undergraduate and graduate students will rely on this resource.

---

Praise for the First Edition

" . . . an excellent addition to an upper-level undergraduate course on environmental statistics, and . . . a 'must-have' desk reference for environmental practitioners dealing with censored datasets."
-Vadose Zone Journal

Statistical Methods for Censored Environmental Data Using Minitab(r) and R, Second Edition introduces and explains methods for analyzing and interpreting censored data in the environmental sciences. Adapting survival analysis techniques from other fields, the book translates well-established methods from other disciplines into new solutions for environmental studies.

This new edition applies methods of survival analysis, including methods for interval-censored data to the interpretation of low-level contaminants in environmental sciences and occupational health. Now incorporating the freely available R software as well as Minitab(r) into the discussed analyses, the book features newly developed and updated material including:
A new chapter on multivariate methods for censored data
Use of interval-censored methods for treating true nondetects as lower than and separate from values between the detection and quantitation limits ("remarked data")
A section on summing data with nondetects
A newly written introduction that discusses invasive data, showing why substitution methods fail
Expanded coverage of graphical methods for censored data

The author writes in a style that focuses on applications rather than derivations, with chapters organized by key objectives such as computing intervals, comparing groups, and correlation. Examples accompany each procedure, utilizing real-world data that can be analyzed using the Minitab(r) and R software macros available on the book's related website, and extensive references direct readers to authoritative literature from the environmental sciences.

Statistics for Censored Environmental Data Using Minitab(r) and R, Second Edition is an excellent book for courses on environmental statistics at the upper-undergraduate and graduate levels. The book also serves as a valuable reference for¿environmental professionals, biologists, and ecologists who focus on the water sciences, air quality, and soil science.

Produktdetails

• Wiley Desktop Editions .
• Verlag: Wiley & Sons
• 2. Aufl.
• Seitenzahl: 344
• Erscheinungstermin: 1. Februar 2012
• Englisch
• Abmessung: 240mm x 161mm x 23mm
• Gewicht: 612g
• ISBN-13: 9780470479889
• ISBN-10: 0470479884
• Artikelnr.: 34160376

Autorenporträt

DENNIS R. HELSEL, PHD, is owner and Principal Scientist of Practical Stats, where he designs and conducts training courses in environmental statistics for scientists. He has over thirty years of experience working with the U.S. Geological Survey and is the author of numerous published articles on nondetect data and statistical methods in the environmental sciences. Dr. Helsel is the recipient of the Distinguished Service Award from the U.S. Department of the Interior (2007) as well as the Distinguished Achievement Award from the American Statistical Association (2003).

Inhaltsangabe

Preface ix Acknowledgments xi Introduction to the First Edition: An
Accident Waiting To Happen xiii Introduction to the Second Edition:
Invasive Data xvii 1 Things People Do with Censored Data that Are Just
Wrong 1 Why Not Substitute--Missing the Signals that Are Present in the
Data 3 Why Not Substitute?--Finding Signals that Are Not There 8 So Why Not
Substitute? 10 Other Common Misuses of Censored Data 10 2 Three Approaches
for Censored Data 12 Approach 1: Nonparametric Methods after Censoring at
the Highest Reporting Limit 13 Approach 2: Maximum Likelihood Estimation 14
Approach 3: Nonparametric Survival Analysis Methods 17 Application of
Survival Analysis Methods to Environmental Data 17 Parallels to Uncensored
Methods 21 3 Reporting Limits 22 Limits When the Standard Deviation is
Considered Constant 23 Insider Censoring-Biasing Interpretations 29
Reporting the Machine Readings of all Measurements 33 Limits When the
Standard Deviation Changes with Concentration 34 For Further Study 36 4
Reporting, Storing, and Using Censored Data 37 Reporting and Storing
Censored Data 37 Using Interval-Censored Data 41 Exercises 42 5 Plotting
Censored Data 44 Boxplots 44 Histograms 46 Empirical Distribution Function
47 Survival Function Plots 49 Probability Plot 52 X-Y Scatterplots 59
Exercises 61 6 Computing Summary Statistics and Totals 62 Nonparametric
Methods after Censoring at the Highest Reporting Limit 62 Maximum
Likelihood Estimation 64 The Nonparametric Kaplan-Meier and Turnbull
Methods 70 ROS: A "Robust" Imputation Method 79 Methods in Excel 86
Handling Data with High Reporting Limits 86 A Review of Comparison Studies
87 Summing Data with Censored Observations 94 Exercises 98 7 Computing
Interval Estimates 99 Parametric Intervals 100 Nonparametric Intervals 103
Intervals for Censored Data by Substitution 103 Intervals for Censored Data
by Maximum Likelihood 104 Intervals for the Lognormal Distribution 112
Intervals Using "Robust" Parametric Methods 125 Nonparametric Intervals for
Censored Data 126 Bootstrapped Intervals 136 For Further Study 140
Exercises 141 8 What Can be Done When All Data Are Below the Reporting
Limit? 142 Point Estimates 143 Probability of Exceeding the Reporting Limit
144 Exceedance Probability for a Standard Higher than the Reporting Limit
148 Hypothesis Tests Between Groups 151 Summary 152 Exercises 152 9
Comparing Two Groups 153 Why Not Use Substitution? 154 Simple Nonparametric
Methods After Censoring at the Highest Reporting Limit 156 Maximum
Likelihood Estimation 161 Nonparametric Methods 167 Value of the
Information in Censored Observations 178 Interval-Censored Score Tests:
Testing Data that Include (DL to RL) Values 180 Paired Observations 183
Summary of Two-Sample Tests for Censored Data 192 Exercises 192 10
Comparing Three or More Groups 194 Substitution Does Not Work--Invasive
Data 195 Nonparametric Methods after Censoring at the Highest Reporting
Limit 196 Maximum Likelihood Estimation 199 Nonparametric Method--The
Generalized Wilcoxon Test 209 Summary 215 Exercises 216 11 Correlation 218
Types of Correlation Coefficients 218 Nonparametric Methods after Censoring
at the Highest Reporting Limit 219 Maximum Likelihood Correlation
Coefficient 224 Nonparametric Correlation Coefficient--Kendall's Tau 227
Interval-Censored Score Tests: Testing Correlation with (DL to RL) Values
230 Summary: A Comparison Among Methods 232 For Further Study 234 Exercises
235 12 Regression and Trends 236 Why Not Substitute? 237 Nonparametric
Methods After Censoring at the Highest Reporting Limit 239 Maximum
Likelihood Estimation 249 Akritas-Theil-Sen Nonparametric Regression 258
Additional Methods for Censored Regression 264 Exercises 266 13
Multivariate Methods for Censored Data 268 A Brief Overview of Multivariate
Procedures 269 Nonparametric Methods After Censoring at the Highest
Reporting Limit 273 Multivariate Methods for Data with Multiple Reporting
Limits 288 Summary of Multivariate Methods for Censored Data 296 14 The
NADA for R Software 297 A Brief Overview of R and the NADA Software 297
Summary of the Commands Available in NADA 300 Appendix: Datasets 303
References 309 Index 321